EP0382581B1 - Bildverarbeitungsvorrichtung - Google Patents

Bildverarbeitungsvorrichtung Download PDF

Info

Publication number
EP0382581B1
EP0382581B1 EP19900301476 EP90301476A EP0382581B1 EP 0382581 B1 EP0382581 B1 EP 0382581B1 EP 19900301476 EP19900301476 EP 19900301476 EP 90301476 A EP90301476 A EP 90301476A EP 0382581 B1 EP0382581 B1 EP 0382581B1
Authority
EP
European Patent Office
Prior art keywords
difference
value
pixel data
target pixel
pixel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
EP19900301476
Other languages
English (en)
French (fr)
Other versions
EP0382581A3 (de
EP0382581A2 (de
Inventor
Yasuhiro Yamada
Hiroshi Tanioka
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Canon Inc
Original Assignee
Canon Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from JP1031409A external-priority patent/JP2683085B2/ja
Priority claimed from JP1031411A external-priority patent/JP2810396B2/ja
Priority claimed from JP1031405A external-priority patent/JP2683084B2/ja
Priority claimed from JP1031404A external-priority patent/JPH02210959A/ja
Priority claimed from JP1031408A external-priority patent/JP2810395B2/ja
Priority claimed from JP1284879A external-priority patent/JPH03147480A/ja
Application filed by Canon Inc filed Critical Canon Inc
Publication of EP0382581A2 publication Critical patent/EP0382581A2/de
Publication of EP0382581A3 publication Critical patent/EP0382581A3/de
Application granted granted Critical
Publication of EP0382581B1 publication Critical patent/EP0382581B1/de
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/40—Picture signal circuits
    • H04N1/40087—Multi-toning, i.e. converting a continuous-tone signal for reproduction with more than two discrete brightnesses or optical densities, e.g. dots of grey and black inks on white paper
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/40—Picture signal circuits
    • H04N1/405—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels
    • H04N1/4051—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels producing a dispersed dots halftone pattern, the dots having substantially the same size
    • H04N1/4052—Halftoning, i.e. converting the picture signal of a continuous-tone original into a corresponding signal showing only two levels producing a dispersed dots halftone pattern, the dots having substantially the same size by error diffusion, i.e. transferring the binarising error to neighbouring dot decisions

Definitions

  • the present invention relates to an image processing apparatus for digitizing image data to binary values or multi-levels and, more particularly, to an image processing apparatus for half-tone processing input image data.
  • multi-level image data of a target pixel is binarized (converted into a highest density level or a lowest density level) and a difference between the binary level and the multi-value image data before the binarization is added with a predetermined weight and the resultant data is added to the data of the pixel near the target pixel.
  • the target pixel is binarized to black or white by using the already binarized data of pixels near the target pixel.
  • Two weighted average values using the respective pixels near the target pixel are obtained, the average of the two average values is set to a threshold value, and the image data of the target pixel is binarized on the basis of the threshold value.
  • the above error diffusion method is of the type in which the difference between the input image data and the output image data is corrected, the densities of the input image and the output image can be preserved, so that an image having excellent resolution and gradation can be provided.
  • the target pixel is merely approximated to the average value of the region including the target pixel and is binarized. Therefore, there are drawbacks such that the number of gradations is limited, a texture of a low frequency which is peculiar to the image having a gentle density change occurs, and the picture quality deteriorates.
  • EP-A-0279419 proposes a binarization arrangement in which an average is calculated of input density values for an area around target pixel.
  • a first threshold is obtained by adding a predetermined value to the average, and if the target pixel has a density which is greater than the first threshold it is binarized to "one".
  • a second threshold value is calculated by subtracting a predetermined value from the average, and if the target pixel has a density value less than the second predetermined value it is binarized to "zero", and if the density of the target pixel is between the two threshold values, it is binarized using a "dither" process.
  • image processing apparatus as set out in claim 1 and an image processing method as set out in claim 19.
  • image processing method as set out in claim 19.
  • An embodiment of the present invention provides an image processing apparatus in which an image having excellent gradation and resolution can be obtained in a short time by a simple hardware construction.
  • Fig. 1(1) is a diagram showing multi-level data of each pixel of an input image.
  • f(i, j) denotes multi-level density data of an input image at the position of a target pixel to be binarized and is set to a normalized value of 0 to 1.
  • the binarizing process has already been finished at pixel positions above a broken line. After the target pixel was binarized, the similar processes are sequentially executed such that f(i, j+1), f(i, j+2), ....
  • Fig. 1(2) is a diagram showing binary image data.
  • B(i, j) denotes a density (having a value of 0 or 1) after the target pixel was binarized.
  • the portion surrounded by a broken line relates to the pixel data which has already been binarized when the target pixel is processed. Those binarized pixel data are used when the target pixel is binarized.
  • Fig. 1(3) is a diagram showing weighting masks.
  • R denotes an example of a weighting mask to obtain an average density and is expressed by a matrix of a size of 3 x 3.
  • the weight at the position corresponding to the target pixel is set to R(0, 0), and R(0, -1) is 0.
  • the average densities of the output images near the target pixel are set to m1(i,j) and m0(i,j) on the assumptions that the target pixel is binarized into black or into white respectively, and are obtained by the following equation.
  • B(i, j) 0, that is, the target pixel is binarized to white
  • the multi-level density f(i,j) of the target pixel is binarized by using the average densities m1 and m0 according to the following equations.
  • E(i,j) denotes a difference which occurs when a multi-level density f(i,j-1) of the pixel (i,j-1) which precedes the target pixel (i,j) by one pixel was binarized to a binary density B(i,j-1). That is, the process such that the input pixel density f(i,j-1) was binarized to 1 or 0 means that the pixel (i,j-1) was approximated to either m1(i,j-1) or m0(i,j-1) as an average density near the pixel (i,j-1).
  • the difference of f(i,j-1) - m1 or f(i,j-1) - m0 occurs with the multi-level density f(i,j-1) of the input image. Therefore, the corrected value, after adding the binarisation difference E(i,j) to the target pixel f(i,j), is binarized, so that density on the image after completion of the binarization can be completely preserved with respect to the whole input image region.
  • the present system has the useful feature that the processes with respect to a point are executed in consideration of such a binarization difference. As compared with the above average density approximating method, the half-tone reproducing capability is remarkably improved.
  • E(i, j+1) denotes a difference which is distributed to the pixel (i, j+1) which is subsequent to the target pixel (i,j) by one pixel.
  • E(i,j+1) is set to a value which is obtained by subtracting m1 from f(i,j) + E(i,j) in the case where f(i,j) + E(i,j) > (m1 + m0)/2.
  • E(i,j+1) is set to a value which is obtained by subtracting m0 from f(i,j) + E(i,j) in the case where f(i,j) + E(i,j) ) ⁇ (m1 + m0)/2.
  • FIG. 4 is a block diagram of an image processing apparatus showing an embodiment of the invention.
  • An input sensor unit A comprises a photoelectric converting element such as a CCD or the like and a drive apparatus for scanning the photoelectric converting element.
  • the input sensor unit A reads and scans an original.
  • the image data of the original which was read by the input sensor unit A is successively transmitted to an A/D converter B.
  • the A/D converter B converts the data of each pixel into 6-bit digital data, thereby digitizing into the data having the gradations of 64 levels.
  • a shading correction and the like to correct a sensitivity variation of the CCD sensor and an illuminance variation by an illuminating light source are executed by digital calculating processes in a correction circuit C.
  • the corrected data is sent to a binarization circuit D.
  • the binarization circuit D the 6-bit multi-level image data which was input is digitized into the 1-bit binary data by the foregoing system.
  • a printer E is constructed by a laser beam system or an ink jet system. On the basis of the binary data sent from the binarization circuit D, the printer E on/off controls the dots and reproduces an image onto a recording paper.
  • Fig. 5 is a block diagram showing the details of the binarizing circuit D in Fig. 4.
  • reference numerals 1 and 2 denote delay RAMs each for storing the binary data which was binarized by an amount of one line; 3 to 7 and 11 indicate D F/F (D flip-flops) each for delaying the binary data by one pixel; 8 an average density calculation ROM for calculating an average density of the pixels around the target pixel and outputting a threshold value; 9 a subtracter for calculating the difference between the multi-level data of the target pixel which was input and the threshold value; 10 a comparator for comparing the threshold value which is output from the ROM 8 with the multi-level data of the target pixel; 11 the D F/F; 12 an ROM for calculating difference data to be added to the multi-level data which is input subsequently to the target pixel; and 13 an adder for adding the input data and the difference data which is output from the ROM 12.
  • D F/F D flip-flops
  • the comparator 10 outputs data B(i,j) of one-bit which was binarized on the basis of the equation 3 to the D F/F 7 and the printer E.
  • the binary data is input to the RAMs 2 and 1 for delaying every line.
  • the binary data B(i-1,j+1) which was delayed by one line by the RAM 2 and the binary data B(i-2,j+1) which was delayed by two lines by the RAM 1 are output to the ROM 8.
  • the D F/F 3 outputs the binary data B(i-2,j) to the ROM 8; the D F/F 4 outputs the B(i-2, j-1); the D F/F 5 outputs the B(i-1,j); D F/F 6 outputs the B(i-1,j-1); and the D F/F 7 outputs the B(i,j-1), respectively, to the ROM 8.
  • the binary data relates to binary images of the peripheral pixel for the input image f(i,j). If the binary images are connected to input addresses in the ROM 8, the binary threshold value can be obtained at a high speed because the binary threshold value (m1(i,j) + m0(i,j))/2 shown in the equation 3 has previously been stored in the ROM 8 on the basis of the equations 1 and 2.
  • the binary threshold value is input to the subtracter 9 and comparator 10.
  • f(i,j) + E(i,j) is input from the D F/F 11 to the subtracter 9 and comparator 10.
  • the subtracter 9 calculates the difference f(i,j) + E(i,j) - (m 1 (i,j) + m 0 (i,j))/2 between both sides of the inequality in the equation 3.
  • the difference value is input to the ROM 12.
  • the comparator 10 compares f(i,j) + E(i,j) and (m1(i,j) + m0(i,j))/2 and outputs the binary data B(i,j).
  • the E(i,j+1) shown in the equation 4 is calculated on the basis of the value of B(i,j) from the comparator 10 and f(i,j) + E(i,j) - (m1(i,j) + m0(i,j))/2 from the subtracter 9.
  • Fig. 6 shows an example of the table stored in the ROM 12.
  • the weight mask 1 shown in Fig. 3 since the weight mask 1 shown in Fig. 3 has been used, in order to normalize to the 6-bit image density levels (0 to 63) which are actually input, the values obtained by the equations 1 and 2 are increased by 63 times and the resultant value is stored as the value which was converted into the 6-bit value into the average density calculation ROM table. In this case, the weight mask 1 is set as shown in Fig. 7.
  • the output f(i,j) + E(i,j) - (m 1 (i,j) + m 0 (i,j))/2 of the subtracter 9 is input as an absolute value into the ROM 12 and the positive/negative sign is determined in accordance with the value of B(i,j).
  • the difference E(i,j+1) obtained by the ROM 12 is added to the input image data f(i,j+1) by the adder 13.
  • the D F/F 11 delays the addition value by a period of time of the data of one clock.
  • the embodiment can be easily realized by merely adding calculation ICs of a few chips as compared with the average density approximating method. Moreover, the gradations can be extremely improved.
  • the average density is calculated on the basis of the binary data and the binarization is executed on the basis of the average density.
  • a processing amount for binarization can be fairly reduced.
  • the gradations can be remarkably improved by correcting the difference between the average density which is generated upon binarization and the input multi-level data.
  • the binarization difference E has been distributed to only the next pixel and corrected in the equation 3. However, if the binarization difference E is distributed to a plurality of pixels in a manner such that 3E(i,j+1)/4 is distributed to the pixel (i,j+1) and E(i,j+1)/4 is distributed to the pixel (i,j+2) in Fig. 1, even if the average processing mask is small, the gradation reproducing capability is improved.
  • the difference E can be also two-dimensionally distributed to a plurality of pixels near the target pixel at a predetermined distribution ratio.
  • the uniform image can be also obtained with respect to the sub scanning direction as well as in the main scanning direction and the reproducibility is improved.
  • Equation 3 In the case of two-dimensionally distributing the difference E in two directions (one adjacent pixel and one pixel existing under the target pixel by one line), the equation 3 is as follows.
  • the difference transfer pixels are set to two pixel positions which are neighboring to the target pixel in two orthogonal directions, that is, two points of (i,j+1) and (i+1,j).
  • the difference value [f(i,j) + E(i,j) - m (i,j)] which is obtained in the embodiment is divided into two values of E1(i,j+1) and E2(i+1,j) and they are respectively added to the input density value, thereby executing the correction.
  • Fig. 8 is a block diagram of a binarizing circuit in the case of two-dimensionally distributing the binarization difference E.
  • the same parts and elements as those in Fig. 5 are designated by the same reference numerals and their descriptions are omitted.
  • Reference numeral 14 denotes a distributor for distributing the difference E which was sent from the ROM 12 into the half portions in order to distribute them to the f(i,j+1) pixel and f(i+1,j) pixel.
  • Reference numeral 15 indicates an adder for adding f(i+1,j) sent from the correction circuit C and E2(i+1,j) sent from the distributor 14.
  • Reference numeral 16 represents an RAM for delaying the data sent from the adder 15 by a period of time corresponding to one line minus one pixel. By providing the RAM 16, the difference generated at the previous line can be added to the target pixel f(i,j).
  • the uniformity can be increased as compared with the case of distributing the difference to one adjacent pixel and the generation of a periodic pattern can be prevented.
  • the directivity does not appear and the edge portion can be clearly reproduced.
  • the weighting mask 1 shown in Fig. 3 has been used as a weighting mask to obtain the average density.
  • the reproducibility of the gradations can be also further improved and the reproducibility of the resolution information which is required in a character original or the like is also improved.
  • the weighting mask is not limited to the above masks, but any mask of a similar form can be also used.
  • the weighting mask has weights which increase as the pixel to be processed approaches the target pixel, its gradient and distribution are not limited. Pixels at spaced apart positions which are not adjacent can be also used.
  • the inevitable dividing process is unnecessary.
  • the difference generated in such a case is eliminated and the densities of the input image and output image can be completely preserved.
  • the difference E(i,j) which is generated when the previous pixel (i,j-1) of the target pixel (i,j) is binarized, is added to f(i,j) of the target pixel. This creates the corrected value which is compared with the average density, thereby executing the binarization.
  • the difference E(i,j) is incorporated in the calculation of the average value when the target pixel is binarized, thereby executing the binarization.
  • m1(i,j) and m0(i,j), obtained by the equations 1 and 2 in the embodiment 1, are set as follows.
  • the multi-level density f(i,j) of the target pixel is binarized by the following equations.
  • Fig. 9 is a block diagram of an image processing apparatus which realizes the embodiment 2.
  • reference numerals 1 to 12 indicate the same component elements as those in the embodiment 1 and their descriptions are omitted.
  • Reference numeral 20 denotes a subtracter to subtract the difference E(i,j) generated at the previous pixel from the average density value sent from the average density calculation ROM 8.
  • the threshold value (m1'(i,j) + m0'(i,j))/2 is obtained by combining the difference E in the subtracter 20 with the average density which was output from the ROM 8, and is input to the subtracter 9 and comparator 10.
  • the multi-level data f(i,j) of the input image is input to the subtracter 9 and comparator 10.
  • the subtracter 9 calculates the difference f(i,j) - (m 1 ′(i,j) + m 0 ′(i,j))/2 between both sides of the inequality in the equation 7.
  • the comparator 10 compares f(i,j) with (m1′(i,j) + m0′(i,j))/2 and outputs the binary data B(i,j).
  • the binarization difference E(i,j+1) is derived by the ROM 12.
  • the difference E(i,j+1) is delayed by the D F/F 11 and is input to the subtracter 20.
  • the characteristic processes of the present system are as follows. As shown in the equations 9 and , in the comparison of the target pixel correction values and the average values m1 and m0 upon binarization, if the target pixel correction value lies within a predetermined (according to the value of ⁇ ) region (namely, the difference E(i,j+1) lies within a predetermined region) in which it has a value near the selected average value m1 or m0, the difference between the selected average value and the target pixel correction value is assigned as the correction value upon binarization of the next pixel in accordance with the equation .
  • the density change of the image near the target pixel is small, so that it is determined such that the image relates to the image area having a half tone. Therefore, the difference between the image density and the average density value generated by binarizing is corrected by the next pixel, thereby enabling the gentle density change of the image to be false half-tone processed with a high fidelity. Namely, the gradations can be improved.
  • the correction value is set to 0 for the pixel in such a case and the deterioration in resolution due to the preservation of the density is suppressed.
  • the density is preserved in the binary image by using the binarization difference in accordance with the image density change and in a high contrast image portion of a character or the like, in order to prevent the blurring of the image due to the preservation of the density, the binarization difference is not used.
  • Fig. 10 is a block diagram showing the details of the binarizing circuit D in the embodiment 3.
  • reference numerals 1 and 2 denote the delay RAMs each for storing the binarized binary data by an amount corresponding to one line; 3 to 7 and 11 indicate the D F/F (D flip-flops) each for delaying the binary data by a period of time corresponding to one line; 8 the average density calculation ROM for calculating the average densities of the pixels around a target pixel and for outputting a threshold value; 9 a subtracter for calculating the difference between the input multi-level data of the target pixel and the threshold value; 10 the comparator for comparing the threshold value which is output from the ROM 8 with the multi-level data of the target pixel; 11 the D F/F; 100 an ROM for calculating difference data to be added to the multi-level data which is input next to the target pixel; and 13 the adder for adding the input data and the difference data which is output from the ROM 100.
  • D F/F D flip-flops
  • the comparator 10 outputs the data B(i,j) of one-bit which was binarized on the basis of the equation 3 to the D F/F 7 and the printer E.
  • the binary data is input to the RAMs 2 and 1 each for delaying every line.
  • the binary data B(i-1, j+1) which was delayed by one line by the RAM 2 is output to the ROM 8.
  • the binary data B(i-2, j+1) which was delayed by two lines by the RAM 1 is output to the ROM 8.
  • the D F/F 3 outputs the B(i-2,j)
  • the D F/F 4 outputs the B(i-2,j-1)
  • the D F/F 5 outputs the B(i-1,j)
  • the D F/F 6 outputs the B(i-1,j-1)
  • the D F/F 7 outputs the B(i,j-1), respectively, to the ROM 8.
  • the binary data relates to the binary images of the peripheral pixels for the input image f(i,j). If the binary images are connected to input addresses in the ROM 8, the binary threshold value can be obtained at a high speed because the binary threshold value (m1(i,j) + m0(i,j))/2 shown in the equation 3 has previously been stored in the ROM 8 on the basis of the equations 1 and 2
  • the threshold value is input to the subtracter 9 and comparator 10.
  • f(i,j) + E(i,j) is input from the D F/F 11 to the subtracter 9 and comparator 10.
  • the subtracter 9 calculates the difference f(i,j) + E(i,j) - (m 1 (i,j) + m 0 (i,j))/2 between both sides of the inequality in the equation 3.
  • the comparator 10 compares f(i,j) + E(i,j) and (m1(i,j + m0(i,j))/2 and outputs the binary data B(i,j).
  • the difference E(i,j+1) is previously calculated and stored into the difference calculation ROM 100. Due to this, the E(i,j+1) is obtained by the table conversion by inputting the binary data B(i,j) and f(i,j) + E(i,j ) - (m 1 (i,j) + m 0 (i,j))/2 as an output of the subtracter 9 into the ROM 12.
  • Fig. 11 shows an example of the table stored in the ROM 100.
  • the weight mask 1 shown in Fig. 3 has been used. Therefore, in order to normalize to the image density levels (0 to 63) of six bits which are actually input, the value obtained by the equations 1 and 2 is increased by 63 times and the resultant values are stored into the average density calculation ROM table as the values which were converted into 6-bit values. In this case, the weight mask 1 is set as shown in Fig. 7.
  • the output f(i,j) + E(i,j) - (m1(i,j) + m0(i,j))/2 of the subtracter 9 is input as an absolute value into the ROM 100.
  • the positive/negative sign is determined in accordance with the value of B(i,j).
  • the difference E(i,j+1) obtained in the ROM 100 is added to the input image data f(i,j+1) by the adder 13.
  • the D F/F 11 delays the addition value by a period of time of one clock of the data.
  • the embodiment 3 can be easily realized by merely adding calculation ICs of a few chips.
  • the average density is calculated on the basis of the data which has already been binarized and the binarization is executed on the basis of the average density, so that a processing amount for binarization can be remarkably reduced. Moreover, when the difference between the average density which is generated upon binarization and the input multi-level data lies within a predetermined range, the difference is corrected, so that the half-tone process having the excellent gradations can be executed.
  • the difference between the average density and the input multi-level data is larger than the predetermined value, the difference is not corrected.
  • the deterioration in resolution due to the preservation of the densities is prevented and the edge portion can be clearly reproduced.
  • the binarization difference E has been distributed to only the next pixel and the correction has been executed in the equation 3.
  • the binarization difference E can be also two-dimensionally distributed to a plurality of pixels near the target pixel at a predetermined distribution ratio.
  • the uniform image can be also obtained with respect to the sub scanning direction as well as the main scanning direction and the reproducibility is improved.
  • the weighting mask has weights which increase as the pixel approaches the target pixel, its gradient and distribution are not limited.
  • the pixels at spaced apart positions which are not adjacent can be also used.
  • the invention can be widely used in image processing apparatuses such as facsimile apparatus, copying machine, and the like.
  • the binarization difference E has been divided into the cases shown by the equations 9 and so that when the difference E is equal to or larger than a predetermined value set using the constant ⁇ , the difference E is set to 0 and is not distributed to the next pixel.
  • the value of the constant ⁇ can be also changed in accordance with the average density value or target pixel density value.
  • ⁇ is set so as to decrease as the average density approaches 0 or 1
  • the edge portion of a black character in the white background, a blank character in the black background, or the like can be more finely binarized.
  • the character portion can be highly finely binarized in a manner similar to the embodiment 2.
  • the weighting mask of the matrix of 3 x 3 shown in Fig. 3 has been used.
  • the half-tone portion can be more smoothly binarized and the edge portion of a character portion or the like can be more finely binarized and reproduced.
  • the calculation of the correction value E for preservation of all densities during the processing has been executed by using the average values m0 and m1.
  • the discrimination or the like to see if the value of E is set to 0 or not in the edge portion or the like can be realized by a well-known technique. For example, a two-dimensional Laplacian is obtained from the image data to be binarized, this value is processed on the basis of the threshold value, the edge portion is determined on the basis of the result of the discrimination of such a threshold value processing, and E is set to 0 in the edge portion. Even by this method, the similar effect is derived.
  • the edge portion is designated in a wide region on the basis of a command obtained by an area designating operation of the operator without switching the processes every pixel and E can be also set to 0 in such a region.
  • the data of the target pixel is binarized by using (m1 + m0)/2 as a threshold value.
  • Fig. 14(1) is a diagram showing a multi-level density of each pixel of the input image.
  • f(i,j) denotes multi-level density data of the input image at the position of the target pixel to be binarized, i.e. set to a normalized value of 0 to 1.
  • the binarization has already been finished at the pixel positions above a broken line. After the target pixel was binarized, the similar binarization is sequentially executed in a manner such that f(i,j+1), f(i,j+2), ....
  • Fig. 14(2) is a diagram showing binary image data.
  • B(i,j) shows the density (having the value of 0 or 1) after the target pixel was binarized.
  • the portion surrounded by a broken line denotes the pixel data which have already been binarized upon processing of the target pixel. Those pixel data are used to binarize the target pixel.
  • Fig. 14(3) is a diagram showing weighting masks.
  • R denotes an example of the weighting mask to obtain the average density and is expressed by a matrix of a size of 3 x 3.
  • the weighted average density of the binary images near the target pixel is set to m(i,j) and is obtained by the following equation.
  • the target pixel f(i,j) is binarized by using the average density m(i,j) and the binary correction value E(i,j) which has already been assigned in accordance with the following equation.
  • Fig. 15 is a diagram showing the equation .
  • E(i,j) denotes a difference which is generated when the multi-level density f(i,j-1) of the pixel which precedes the target pixel (i,j) by one pixel, that is, of the pixel (i,j-1) was binarized to the binary density B(i,j-1).
  • E(i,j) corresponds to the difference value between the multi-level density f(i,j-1) and the average density m(i,j-1) of the pixel near the target pixel.
  • the reproducing capability of the half tone is remarkably improved as compared with that in the average density approximating method.
  • E(i,j+1) denotes a difference which is distributed to the pixel f(i,j+1) which is located beyond the target pixel (i,j) by one pixel.
  • the image reproducing capability is equal to or higher than that in the error diffusion method, because in spite of the fact that the difference is merely corrected by one adjacent pixel, an effect similar to that in the case of distributing the difference to a plurality of pixels and correcting is equivalently obtained by obtaining the average density by using a plurality of data after completion of the binarization.
  • Fig. 16 is a block diagram showing the details of the binarizing circuit D in the embodiment 7.
  • reference numerals 101 and 102 denote delay RAMs each for storing the binarized data by an amount of one line; 103 to 107 and 111 indicate D F/F (D flip-flops) each for delaying the binary data by a period of time of one pixel; 108 an average density calculation ROM for calculating the average density of a predetermined region from the binary data of the pixels around the target pixel and for outputting the average density as a threshold value when the data of the target pixel is binarized; 109 a subtracter for calculating the difference between the input multi-level data of the target pixel and the threshold value which is output from the ROM 108; 110 a comparator for comparing the threshold value which is output from the ROM 108 with the multi-level data of the target pixel; and 112 an adder for adding the difference data which is output from the subtracter 109 and the 6-bit multi-level data sent from the correction circuit.
  • D F/F D flip-flops
  • the comparator 110 outputs the 1-bit data B(i,j) which was binarized on the basis of the equation .
  • the binary data is input to the RAMs 102 and 101 for delaying it every line.
  • the binary data B(i-1,j+1) which was delayed by one line by the RAM 102 is output to the ROM 108.
  • the binary data B(i-2,j+1) which was delayed by two lines by the RAM 101 is output to the ROM 108.
  • the D F/F 103 outputs the B(i-2,j)
  • the D F/F 104 outputs the B(i-2,j-1)
  • the D F/F 105 outputs the B(i-1,j)
  • the D F/F 106 outputs the B(i-1,j-1)
  • the D F/F 107 outputs the B(i,j-1), respectively, to the ROM 108.
  • the binary data denote the binary images of the peripheral pixels for the input image f(i,j) as shown in Fig. 14.
  • the binarization threshold value can be obtained at a high speed because the binarization threshold value m(i,j) shown in the equation has previously been stored in the ROM 108 on the basis of the equation .
  • the threshold value is input to the subtracter 109 and comparator 110.
  • f(i,j) + E(i,j) is input from the D F/F 111 to the subtracter 109 and comparator 110.
  • the subtracter 109 calculates the difference between both sides of the inequality in the equation , that is, the difference between the average density value m(i,j) and the input data.
  • the comparator 110 compares f(i,j) + E(i,j) and m(i,j) on the basis of the two inputs and outputs the binary data B(i,j). Then, on the basis of the equation , the difference E(i,j+1) which is output from the subtracter 109 is added to the input image data f(i,j+1) by the adder 112.
  • the D F/F 111 delays the addition value by a period of time of one clock of the data.
  • Fig. 17 shows examples of the weighting masks.
  • the weighting mask of Fig. 17A is used to obtain the average value from the binary data of seven pixels.
  • the weighting mask of Fig. 17B is used to obtain the average value from the binary data of twelve pixels.
  • the weighting mask shown in Fig. 17A is used, in order to normalize to the image density levels (0 to 63) of six bits which are actually input, the values obtained by the equation are increased by 63 times and the resultant values are stored into the average density calculation ROM table as the values which were converted into the 6-bit values.
  • the average density is calculated by using only the binary data which was binarized and the input multi-level data is binarized by using the average density as a threshold value. Therefore, a processing amount for binarization can be reduced as compared with that in the average density approximating method. Moreover, since the difference between the input multi-level data and the average density which is generated when the input multi-level data was binarized is corrected the gradations can be extremely improved.
  • the binarization difference E has been distributed to only the next pixel and corrected, if the binarization difference E is distributed to a plurality of pixels in the main scanning direction in a manner such that, for instance, in Fig. 14, 3E(i,j+1)/4 is distributed to the pixel (i,j+1) and E(i,j+1)/4 is distributed to the pixel (i,j+2), the reproducing capability of the gradations is improved even when the average processing mask is small.
  • the difference E can be also two-dimensionally distributed to a plurality of pixels near the target pixel at a predetermined distribution ratio in a manner similar to the case of the error diffusion method.
  • the uniform image can be obtained in the sub scanning direction as well as the main scanning direction and the reproducibility is improved.
  • the weighting mask has weights which increase as the pixel approaches the target pixel, its gradient and distribution are not limited.
  • the pixels at spaced apart positions which are not adjacent can be also used.
  • the weighting mask of a matrix of 3 x 3 such as shown in an example of Fig. 17A has been used.
  • the calculation of the average density m has easily been realized by the ROM table.
  • such a calculation can be also realized even by using seven AND gates and a plurality of adders.
  • the processing speed can be further made high.
  • by assembling such a processing circuit into a gate array or the like a hardware scale can be remarkably reduced.
  • the correction has been performed by directly adding the difference E(i,j) to the target pixel density f(i,j).
  • the difference E(i,j) is accounted for by subtracting it from the average density by using a subtracter 113, the similar effect can be also obtained.
  • the embodiment 8 relates to a partially modified form of the embodiment 7.
  • the difference E(i,j+1) when the next pixel is binarized is expressed by the following equation.
  • the corrected target pixel density lies within a predetermined range of a value near the average density m (that is, when the difference between the average density m and the corrected target pixel density lies within a predetermined range)
  • the difference between the corrected target pixel density and the average density is assigned as a correction value when the next pixel is binarized in accordance with the equation .
  • the correction value is set to 0 and the correction upon binarization of the next pixel is not executed.
  • the density change of the image near the target pixel is small, so that it is determined that such as image is in the image area having a half-tone. Therefore, the difference between the image density and the average density which occurs due to the binarization is corrected by the next pixel, so that the gentle density change of the image can be false half-tone processed with a high fidelity. That is, the gradations can be improved.
  • the difference is not corrected, there is the edge portion in a character, a diagram, or the like, that is, the target pixel is determined to be suddenly changed as compared with the density of the image near the target pixel.
  • the correction value is set to 0 for the pixel in such a case and the deterioration in resolution due to the preservation of the density is suppressed.
  • the density is preserved in the binary image by using the binarization difference in accordance with the image density change.
  • the correction of the binarization difference is not used.
  • Fig. 19 is a block diagram showing the details of the binarizing circuit D in the embodiment 8.
  • Reference numeral 115 denotes a comparator for comparing the difference which is sent from the subtracter 109, between the multi-level data of the target pixel and the threshold value, with a predetermined value ( ⁇ ).
  • Reference numeral 116 indicates a selector for selecting either 0 or an output of the subtracter 109 on the basis of a select signal from the comparator 115.
  • the comparator 110 outputs the 1-bit data B(i,j) which was binarized on the basis of the equation .
  • the binary data is input to the RAMs 102 and 101 each for delaying it every line.
  • the binary data B(i-1,j+1) which was delayed by one line by the RAM 102 is output to the ROM 108.
  • the binary data B(i-2,j+1) which was delayed by two lines by the RAM 101 is output to the ROM 108.
  • the D F/F 103 outputs the B(i-2,j)
  • the D F/F 104 outputs the B(i-2,j-1)
  • the D F/F 105 outputs the B(i-1,j)
  • the D F/F 106 outputs the B(i-1, j-1)
  • the D F/F 107 outputs the B(i,j-1), respectively, to the ROM 108.
  • the binary data denotes the binary images of the peripheral pixels for the input image f(i,j).
  • the binarization threshold value can be obtained at a high speed because the binarization threshold value m(i,j) shown in the equation has previously been stored in the ROM 108 on the basis of the equation .
  • the threshold value is input to the subtracter 109 and comparator 110.
  • f(i,j) + E(i,j) is input from the D F/F 111 to the subtracter 109 and comparator 110.
  • the subtracter 109 calculates the difference between both sides of the inequality in the equation .
  • the comparator 110 compares f(i,j) + E(i,j) with m(i,j) on the basis of the two inputs and outputs the binary data B(i,j). Then, the difference E(i,j+1) which is output from the subtracter 109 is input to the selector 116 and the comparator 115 on the basis of the equation .
  • the comparator 115 compares the difference E(i,j+1) with the constant ⁇ by the equation and outputs the select signal to the selector 116 on the basis of the result of the comparison.
  • the difference E(i,j+1) is added to the input image data f(i,j+1) by the adder 112.
  • the D F/F 111 delays the addition value by a period of time of one clock of the data.
  • the weighting mask shown in Fig. 17 is used.
  • the average density is calculated by using only the binary data which was binarized and the input multi-level data is binarized by using the average density as a threshold value. Therefore, a processing amount for binarization can be reduced as compared with that in the average density approximating method. Moreover, when the difference between the average density which is generated upon binarization of the input multi-level data and the input multi-level data lies within a predetermined range, the difference is corrected, so that the gradations can be remarkably improved.
  • the difference between the average density and the input multi-level data is larger than a predetermined value, the difference is not corrected.
  • the deterioration in resolution due to the preservation of the density is prevented and the edge portion can be clearly reproduced.
  • the binarization difference E has been distributed to only the next pixel and corrected.
  • the binarization difference E is distributed to a pluraltiy of pixels in the main scanning direction in a manner such that, for instance, in Fig. 14, 3E(i,j+1)/4 is distributed to the pixel (i,j+1) and E(i,j+1)/4 is distributed to the pixel (i,j+2), the reproducing capability of the gradations is improved even if the average processing mask is small.
  • the binarization difference can be also two-dimensionally distributed to a plurality of pixels near the target pixel at a predetermined distribution ratio in a manner similar to the case of the error diffusion method.
  • the uniform image can be obtained in the sub scanning direction as well as the main scanning direction and the reproducibility is improved.
  • the weighting mask has weights which increase as the pixel approaches to the target pixel, its gradient and distribution are not limited.
  • the pixels existing at spaced apart positions which are not adjacent can be also used.
  • the binarization difference E has been divided into two cases shown by the equations and and when the difference E is equal to or larger than a predetermined value set using the constant ⁇ , the difference E is set to 0 and is not distributed to the next pixel.
  • the value of the constant a can be also changed in accordance with the average density value or target pixel density value.
  • ⁇ is set to be reduced as the average density approaches 0 or 1
  • the edge portion of a black character in the white background, a blank character in the black background, or the like can be more finely binarized.
  • the character portion can be finely binarized in a manner similar to the embodiment 9.
  • the weighting mask of a matrix of 3 x 3 as shown in an example of Fig. 17A has been used.
  • the calculation of the average density m has easily been realized by the ROM table.
  • such calculation can be also realized by using seven AND gates and a plurality of adders. In this case, the processing speed can be further made high.
  • a hardware scale can be remarkably reduced.
  • the average density m which is used in the present system has been used in the calculation of the correction value E for preservation of all densities during the processing.
  • the determination or the like to see if E is set to 0 or not in the edge portion or the like can be also realized by the well-known technique. For instance, a two-dimensional Laplacian is obtained from the image data to be binarized, this value is processed to a threshold value, the edge portion is discriminated on the basis of the result of the discrimination of such a threshold value process, and E is set to 0 in the edge portion. The similar result is also obtained even by such a method.
  • the edge portion is designated in a wide area on the basis of a command obtained by the area designating operation by the operator without switching the process every pixel and E can be also set to 0 in such an area.
  • the invention can be also applied to a color image by setting the input data to three colors of R, G, and B.
  • an image having excellent gradation and resolution can be obtained in a short time by a simple hardware construction.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Discrete Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Facsimile Image Signal Circuits (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Claims (20)

  1. Bildverarbeitungseinrichtung mit:
    einer Eingabeeinrichtung (A, B, C) zum Eingeben von Bildelementedaten und
    einer Verarbeitungseinrichtung (1 bis 8, 10; 101 bis 108, 110) für das Verarbeiten von eingegebenen Bildelementedaten zu digitalisierten Ausgabe-Bildelementedaten,
    wobei die Verarbeitungseinrichtung
    eine Digitalisiereinrichtung (10; 110) zum Digitalisieren der eingegebenen Bildelementedaten durch Vergleich mit einem Schwellenwert und
    eine Recheneinrichtung (1 bis 8; 101 bis 108) zum Berechnen eines mittleren Bildelementedatenwertes für einen Bildbereich aufweist, der eine Vielzahl von Bildelementen enthält, die vor einem bestimmten Bildelement (Ziel-Bildelement) digitalisiert sind, um dadurch den Schwellenwert für den Einsatz bei dem Ziel-Bildelement zu erhalten,
    dadurch gekennzeichnet,
    daß die Recheneinrichtung (1 bis 8; 101 bis 108) den mittleren Bildelementedatenwert aus den von der Digitalisiereinrichtung (10, 110) ausgegebenen digitalisierten Bildelementedaten berechnet und
    daß die Bildverarbeitungseinrichtung eine Korrektureinrichtung (9, 11, 12, 13, 14, 15, 16, 100; 109, 111, 112, 113, 115, 116, 120) zum Korrigieren der Funktion der Verarbeitungseinrichtung an mindestens einem Bildelement, welches nach dem Ziel-Bildelement digitalisiert ist, hinsichtlich der Differenz (Digitalisierungsdifferenz) zwischen (i) einem Wert für das Ziel-Bildelement vor dessen Digitalisierung durch die Digitalisiereinrichtung und (ii) einem Bezugswert aufweist, der ein Mittelwert von digitalisierten Bildelementedaten ist, die aus der Digitalisiereinrichtung für eine Bildbereich ausgegeben sind, der eine Vielzahl von vor dem Ziel-Bildelement digitalisierten Bildelementen enthält.
  2. Einrichtung nach Anspruch 1, in der der Bezugswert der Korrektureinrichtung von dem durch die Recheneinrichtung berechneten mittleren Bilddatenwert verschieden ist.
  3. Einrichtung nach Anspruch 2, in der der Bezugswert der Korrektureinrichtung ein mittlerer Bildelementedatenwert für einen Bildbereich ist, der das Ziel-Bildelement enthält, und der durch die Recheneinrichtung berechnete mittlere Bildelementedatenwert der Wert für einen Bildbereich ist, der nicht das Ziel-Bildelement enthält (Fig. 5, 8, 9, 10).
  4. Einrichtung nach Anspruch 3, in der die Korrektureinrichtung eine erste Einrichtung (9), die die Differenz zwischen einem Wert für das Ziel-Bildelement vor dessen Digitalisierung und dem durch die Recheneinrichtung gelieferten Schwellenwert ermittelt, und eine zweite Einrichtung (12; 100) aufweist, die den Differenzwert aus der ersten Einrichtung (9) und die digitalisierten Bildelementedaten für das Ziel-Bildelement aus der Digitalisiereinrichtung (10) aufnimmt und als Digitalisierungsdifferenz einen unter Berücksichtigung der digitalisierten Bildelementedaten für das Ziel-Bildelement revidierten Differenzwert abgibt.
  5. Einrichtung nach Anspruch 4, in der die zweite Einrichtung (12; 100) einen Speicher aufweist, der als Adresseneingangssignale den Differenzwert aus der ersten Einrichtung (9) und die digitalisierten Bildelementedaten aus der Digitalisiereinrichtung (10) aufnimmt.
  6. Einrichtung nach Anspruch 1, in der der Bezugswert der Korrektureinrichtung der gleiche ist wie der durch die Recheneinrichtung berechnete Bildelementedatenwert.
  7. Einrichtung nach Anspruch 6, in der der Bezugswert der Korrektureinrichtung und der durch die Recheneinrichtung berechnete mittlere Bildelementedatenwert beide ein mittlerer Bildelementedatenwert für einen Bildbereich sind, der nicht das Ziel-Bildelement enthält (Fig. 16, 18, 19, 21).
  8. Einrichtung nach Anspruch 7, in der die Korrektureinrichtung eine erste Einrichtung (109) zum Ermitteln der Differenz zwischen einem Wert für das Ziel-Bildelement vor dessen Digitalisierung und dem durch die Recheneinrichtung gelieferten Schwellenwert aufweist.
  9. Einrichtung nach einem der vorangehenden Ansprüche, in der die Korrektureinrichtung die Funktion der Verarbeitungseinrichtung unabhängig von dem Wert der Digitalisierungsdifferenz korrigiert.
  10. Einrichtung nach einem der Ansprüche 1 bis 8, in der die Korrektureinrichtung die Funktion der Verarbeitungseinrichtung hinsichtlich der Digitalisierungsdifferenz für das Ziel-Bildelement nur dann korrigiert, wenn die Digitalisierungsdifferenz für das Ziel-Bildelement nicht einen vorbestimmten Wert übersteigt.
  11. Einrichtung nach Anspruch 4 oder Anspruch 5, in der die zweite Einrichtung (100) als Digitalisierungsdifferenz den revidierten Differenzwert nur dann abgibt, wenn der Differenzwert aus der ersten Einrichtung (9) nicht einen vorbestimmten Wert übersteigt, und andernfalls als Digitalisierungsdifferenz "0" abgibt.
  12. Einrichtung nach Anspruch 8, in der die Korrektureinrichtung eine zweite Einrichtung (115, 116, Fig. 19) aufweist, die den Differenzwert aus der ersten Einrichtung aufnimmt und diesen als Digitalisierungsdifferenz nur dann abgibt, wenn er nicht einen vorbestimmten Wert übersteigt, und anderenfalls als Digitalisierungsdifferenz "0" abgibt.
  13. Einrichtung nach einem der vorangehenden Ansprüche, in der die Korrektureinrichtung eine Einrichtung (11, 13; 15, 16; 111, 112) enthält, die die aus der Eingabeeinrichtung (A, B, C) eingegebenen Bildelementedaten mit der Digitalisierungsdifferenz kombiniert, um korrigierte Eingabe-Bildelementedaten zu erhalten, und die die korrigierten Eingabe-Bildelementedaten zur Digitalisierung zu der Digitalisiereinrichtung weitergibt, um dadurch die Funktion der Verarbeitungseinrichtung zu korrigieren (Fig. 5, 8, 10, 16, 19, 21).
  14. Einrichtung nach einem der Ansprüche 1 bis 12, in der die Korrektureinrichtung eine Einrichtung (20; 119) enthält, die den durch die Recheneinrichtung berechneten mittleren Bildelementedatenwert mit der Digitalisierungsdifferenz kombiniert, um einen abgeänderten Mittelwert zu erhalten, und den abgeänderten Mittelwert als Schwellenwert für die Digitalisierung zu der Digitalisiereinrichtung weitergibt (Fig. 9, 18).
  15. Einrichtung nach einem der vorangehenden Ansprüche, in der die Recheneinrichtung einen Rechenspeicher (8; 108) enthält, der als Adresseneingangssignale die digitalisierten Bildelementedaten für die Bildelemente des Bildbereiches aufnimmt und den mittleren Bildelementedatenwert für den Bildbereich abgibt.
  16. Einrichtung nach einem der vorangehenden Ansprüche, in der die Recheneinrichtung den mittleren Bildelementedatenwert als einen entsprechend einer vorbestimmten Gewichtungsmaske gewichteten Mittelwert berechnet.
  17. Einrichtung nach einem der vorangehenden Ansprüche, in der die Recheneinrichtung eine Einrichtung (1 bis 7; 101 bis 107) enthält, welche die von der Digitalisiereinrichtung abgegebenen digitalisierten Bildelementedaten aufnimmt und verzögert.
  18. Einrichtung nach einem der vorangehenden Ansprüche, in der die Digitalisiereinrichtung (10; 110) die eingegebenen Bildelementedaten binär digitalisiert.
  19. Bildverarbeitungsverfahren, welches das Digitalisieren von eingegebenen-Bildelementedaten im Falle des Digitalisierens eines bestimmten Bildelementes (Ziel-Bildelementes) in Bezug auf einen mittleren Bildelementedatenwert für einen Bildbereich umfaßt, der eine Vielzahl von vor dem ziel-Bildelement digitalisierten Bildelementen enthält,
    dadurch gekennzeichnet, daß der mittlere Bildelementedatenwert ein Mittelwert der digitalisierten Werte für die Bildelemente des Bildbereiches ist und
    daß die Digitalisierung von mindestens einem Bildelement, welches nach dem Ziel-Bildelement digitalisiert wird, hinsichtlich der Differenz zwischen einem Datenwert für das Ziel-Bildelement vor der Digitalisierung und einem Mittelwert der digitalisierten Werte der Bildelemente eines Bildbereiches korrigiert wird, der eine Vielzahl von vor dem Ziel-Bildelement digitalisierten Bildelementen enthält.
  20. Verfahren nach Anspruch 19, bei dem die Bildelementedaten optische Dichtedaten sind.
EP19900301476 1989-02-10 1990-02-12 Bildverarbeitungsvorrichtung Expired - Lifetime EP0382581B1 (de)

Applications Claiming Priority (12)

Application Number Priority Date Filing Date Title
JP31405/89 1989-02-10
JP1031411A JP2810396B2 (ja) 1989-02-10 1989-02-10 画像処理装置
JP31404/89 1989-02-10
JP1031405A JP2683084B2 (ja) 1989-02-10 1989-02-10 画像処理装置
JP1031404A JPH02210959A (ja) 1989-02-10 1989-02-10 画像処理装置
JP31409/89 1989-02-10
JP1031408A JP2810395B2 (ja) 1989-02-10 1989-02-10 画像処理装置
JP1031409A JP2683085B2 (ja) 1989-02-10 1989-02-10 画像処理装置
JP31408/89 1989-02-10
JP31411/89 1989-02-10
JP284879/89 1989-11-02
JP1284879A JPH03147480A (ja) 1989-11-02 1989-11-02 画像処理装置

Publications (3)

Publication Number Publication Date
EP0382581A2 EP0382581A2 (de) 1990-08-16
EP0382581A3 EP0382581A3 (de) 1991-10-16
EP0382581B1 true EP0382581B1 (de) 1996-05-08

Family

ID=27549596

Family Applications (2)

Application Number Title Priority Date Filing Date
EP19900301475 Expired - Lifetime EP0382580B1 (de) 1989-02-10 1990-02-12 Bildverarbeitungsgerät
EP19900301476 Expired - Lifetime EP0382581B1 (de) 1989-02-10 1990-02-12 Bildverarbeitungsvorrichtung

Family Applications Before (1)

Application Number Title Priority Date Filing Date
EP19900301475 Expired - Lifetime EP0382580B1 (de) 1989-02-10 1990-02-12 Bildverarbeitungsgerät

Country Status (2)

Country Link
EP (2) EP0382580B1 (de)
DE (2) DE69027870T2 (de)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0395404B1 (de) * 1989-04-27 1999-12-01 Canon Kabushiki Kaisha Bildverarbeitungsvorrichtung
US5515180A (en) * 1992-11-24 1996-05-07 Sharp Kabushiki Kaisha Image processing device
EP0670653B1 (de) * 1994-03-02 2001-07-11 Hewlett-Packard Company, A Delaware Corporation Erzeugung von Mehrfachtonbildern
EP0670654B1 (de) * 1994-03-02 2001-04-18 Hewlett-Packard Company, A Delaware Corporation Erzeugung von Mehrfachtonbildern
DE69518578T2 (de) * 1994-05-18 2001-04-26 Sharp Kk Kartenartige Kamera mit Bildverarbeitungsfunktion
DE69526559T2 (de) * 1994-11-08 2002-08-22 Xerox Corp Halbtonrasterung von Bildelementpaaren für einen Drucker mit hoher Schärfe
EP0893910A1 (de) * 1997-07-21 1999-01-27 SYFAL S.p.A. Verfahren zur Reproduktion von hochauflösenden Bildern
US7268919B2 (en) 2002-01-17 2007-09-11 Seiko Epson Corporation Image data processing apparatus, method, and program that diffuses gradiation error for each pixel in target block

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS57104369A (en) * 1980-12-22 1982-06-29 Toshiba Corp Binary device for picture with contrast
JPS60157375A (ja) * 1983-12-21 1985-08-17 Fuji Xerox Co Ltd 中間調表現方式
JPS60214160A (ja) * 1984-04-09 1985-10-26 Ricoh Co Ltd 画信号２値化方式
DE3583474D1 (de) * 1984-07-25 1991-08-22 Matsushita Electric Ind Co Ltd Bildsignalverarbeitungsgeraet.
EP0248616A3 (de) * 1986-06-02 1990-03-07 Kabushiki Kaisha Toshiba Bilddruckgerät
US4972502A (en) * 1987-02-17 1990-11-20 Sharp Kabushiki Kaisha Image signal binary encoder

Also Published As

Publication number Publication date
DE69027870D1 (de) 1996-08-29
EP0382580B1 (de) 1996-07-24
EP0382581A3 (de) 1991-10-16
EP0382581A2 (de) 1990-08-16
EP0382580A3 (de) 1991-10-02
DE69027870T2 (de) 1997-01-09
DE69026846T2 (de) 1997-01-09
EP0382580A2 (de) 1990-08-16
DE69026846D1 (de) 1996-06-13

Similar Documents

Publication Publication Date Title
US6118547A (en) Image processing method and apparatus
US5121447A (en) Apparatus for performing gradation processing on image data
US4958218A (en) Image processing method and apparatus with dot-processing
EP0781034B1 (de) Bildverarbeitungsgerät und -verfahren
US5243445A (en) Picture processing apparatus
US5289294A (en) Image processing apparatus
US5157741A (en) Image processing method and apparatus for out-putting dot-processed data with suppression of false contours and other noise
US5577136A (en) Image processing apparatus
US6233360B1 (en) Method and system for hybrid error diffusion processing of image information using adaptive white and black reference values
EP0382581B1 (de) Bildverarbeitungsvorrichtung
US6134355A (en) Binarization using a local average, and using error diffusion
US5760918A (en) Image processing apparatus with conversion and reconversion of the number of bits per pixel
JP2683085B2 (ja) 画像処理装置
JP2801195B2 (ja) 画像処理装置
JP2683084B2 (ja) 画像処理装置
JP2662401B2 (ja) 画像処理装置
JPH01115272A (ja) 画像処理装置
JP2810395B2 (ja) 画像処理装置
JP2848566B2 (ja) 画像処理装置
JP2810396B2 (ja) 画像処理装置
JP2675792B2 (ja) 画像処理装置
JP2662402B2 (ja) 画像処理装置
JP2900907B2 (ja) 画像処理装置
JP2644491B2 (ja) 画像処理装置
JP2848567B2 (ja) 画像処理装置

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): DE FR GB

17P Request for examination filed

Effective date: 19901231

PUAL Search report despatched

Free format text: ORIGINAL CODE: 0009013

AK Designated contracting states

Kind code of ref document: A3

Designated state(s): DE FR GB

17Q First examination report despatched

Effective date: 19940201

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): DE FR GB

REF Corresponds to:

Ref document number: 69026846

Country of ref document: DE

Date of ref document: 19960613

ET Fr: translation filed
PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed
REG Reference to a national code

Ref country code: GB

Ref legal event code: IF02

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20080226

Year of fee payment: 19

Ref country code: DE

Payment date: 20080229

Year of fee payment: 19

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FR

Payment date: 20080221

Year of fee payment: 19

GBPC Gb: european patent ceased through non-payment of renewal fee

Effective date: 20090212

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST

Effective date: 20091030

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20090901

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GB

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20090212

Ref country code: FR

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20090302